Marvel Comic Presentation

Eric Hogue

8/10/2021

Marvel’s Next Breakthrough Characters

Building on over 80 years of success, Marvel is looking to introduce 2 new characters to captivate a new generation of audiences.

Marvel’s Most Succesful Characters

Number of Appearances

  • Issues with characters that were introduced earlier in time.

Appearances over Time

  • Measuring using first appearnace

Marvel’s Top 10 Characters by Total Appearance

Marvel’s Top 10 Characters by Total Appearance

Name Total Appearances Date of 1st Appearance Year of 1st Appearance
Spider-Man (Peter Parker) 4043 Aug-62 1962
Captain America (Steven Rogers) 3360 Mar-41 1941
Wolverine (James "Logan" Howlett) 3061 Oct-74 1974
Iron Man (Anthony "Tony" Stark) 2961 Mar-63 1963
Thor (Thor Odinson) 2258 Nov-50 1950
Benjamin Grimm (Earth-616) 2255 Nov-61 1961
Reed Richards (Earth-616) 2072 Nov-61 1961
Hulk (Robert Bruce Banner) 2017 May-62 1962
Scott Summers (Earth-616) 1955 Sep-63 1963
Jonathan Storm (Earth-616) 1934 Nov-61 1961

Marvel’s Top 10 Character Appearance per Year

Name Total Appearances Average Appearance Per Year Date of 1st Appearance Year of 1st Appearance
Spider-Man (Peter Parker) 4043 68.53 Aug-62 1962
Wolverine (James "Logan" Howlett) 3061 65.13 Oct-74 1974
Iron Man (Anthony "Tony" Stark) 2961 51.05 Mar-63 1963
Captain America (Steven Rogers) 3360 42.00 Mar-41 1941
Benjamin Grimm (Earth-616) 2255 37.58 Nov-61 1961
Reed Richards (Earth-616) 2072 34.53 Nov-61 1961
Hulk (Robert Bruce Banner) 2017 34.19 May-62 1962
Scott Summers (Earth-616) 1955 33.71 Sep-63 1963
Ororo Munroe (Earth-616) 1512 32.87 May-75 1975
Jonathan Storm (Earth-616) 1934 32.23 Nov-61 1961

What Factors Make a Great Character?

Are there certain factors that influence appreances?

The dataset offers information on various characteristics:

-Sex

-Gender/Sexuality

-Eyes

-Hair

-Good/Evil/Neutral Alignment

-Life Status

Summary of Individual Factors

Number of Distinct Characteristics
Sex 4
Alignment 3
Gender/Sexuality 7
Alive/Deceased 2
Hair Color 26
Eye Color 23

Summary of Individual Factors (Sex %)

Category Percent
Agender Characters 0.33
Female Characters 24.41
Genderfluid Characters 0.02
Male Characters 75.25

Summary of Individual Factors (Good/Bad Alignment %)

Category Percent
Bad Characters 49.20
Good Characters 34.51
Neutral Characters 16.29

Summary of Individual Factors (Gender/Sexuality %)

Category Percent
Bisexual Characters 0.16
Genderfluid Characters 0.01
Homosexual Characters 0.43
Pansexual Characters 0.01
Transgender Characters 0.01
Transvestites 0.01
Unclassified 99.38

Summary of Individual Factors

Factors ranked by Completeness:

  1. SEX (100%)

  2. ALIGNMENT (100%)

  3. ALIVE (100%)

  4. HAIR (80%)

  5. EYE (45%)

  6. Gender/Sexuality (<1%)

Summary Stats

Avg # of Appearances Avg # of Appearances/Year Standard Deviation Number of Characters Max # of Appearances Min # of Appearances
19.85 0.53 107.4 12193 4043 1

Summary of Individual Factors

Summary Stats by Sex
SEX Avg # of Appearances Avg # of Appearances/Year Standard Deviation Number of Characters Max # of Appearances Min # of Appearances
Agender Characters 20.12 0.84 54.92 40 348 1
Female Characters 22.83 0.70 88.85 2976 1713 1
Genderfluid Characters 282.50 4.72 352.85 2 532 33
Male Characters 18.82 0.47 112.81 9175 4043 1

Summary of Individual Factors

Summary Stats by Good/Bad Alignment
ALIGN Avg # of Appearances Avg # of Appearances/Year Standard Deviation Number of Characters Max # of Appearances Min # of Appearances
Bad Characters 8.64 0.26 26.42 5999 721 1
Good Characters 35.61 0.91 161.44 4208 4043 1
Neutral Characters 20.31 0.54 112.24 1986 3061 1

Appearance Distribution

Appearance Distribution

Appearance Distribution

Appearance Distribution

Appearance Distribution

Appearance Distribution by Factors (by Sex & Decade)

Apperance Distribution by Factors (By Sex)

Apperance Distribution by Factors (By Align)

What Factors influence Appearances?

Bayesian Regression Analysis

  • See distributions of each individual factor’s R-squared

Mixed Model

  • Determine which factors may interact with each other

  • Fix time variable

Regression Analysis (R-Squared Analsis)

Regression Analysis (Factor Analsis)

(Alignment/Sex/Hair)

ANOVA Table for Align/Sex/Hair
npar Sum Sq Mean Sq F value
ALIGN 3 1059.10567 353.035223 237.129279
SEX 3 163.80595 54.601984 36.675459
HAIR 25 1034.28533 41.371413 27.788653
ALIGN:SEX 5 40.73980 8.147960 5.472881
ALIGN:HAIR 41 191.41171 4.668578 3.135825
SEX:HAIR 23 82.63366 3.592768 2.413217
ALIGN:SEX:HAIR 37 70.67366 1.910099 1.282989

Regression Analysis (Factor Analsis)

(Alignment/Sex/Eye) {.smaller}

ANOVA Table for Align/Sex/Eye
npar Sum Sq Mean Sq F value
ALIGN 3 1056.12050 352.040168 267.496732
SEX 3 161.03693 53.678977 40.787820
EYE 22 3231.43470 146.883396 111.608935
ALIGN:SEX 5 28.78541 5.757082 4.374503
ALIGN:EYE 40 93.43140 2.335785 1.774840
SEX:EYE 29 74.88123 2.582111 1.962010
ALIGN:SEX:EYE 35 91.49578 2.614165 1.986366

Regression Analysis

ANOVA Table for Eye/Hair
npar Sum Sq Mean Sq F value
EYE 23 3782.9077 164.474250 121.650156
HAIR 25 233.0951 9.323804 6.896169
EYE:HAIR 175 407.9287 2.331021 1.724094

Predicting The Next Great Marvel Characters! (Boosting Method)

Gradient Boosting Model

  • Multiple factor variables make predition with traditional regression difficult

  • Fairly predictive

  • Runs efficiently

Top Predicted Characteristics

Summary Stats for Predicted Appearances
SEX ALIGN EYE HAIR GSM pred_appearance_avg
Female Characters Neutral Characters White Eyes Silver Hair Bisexual Characters 84
Female Characters Neutral Characters Green Eyes Black Hair Bisexual Characters 82
Female Characters Good Characters Blue Eyes Blond Hair Bisexual Characters 81
Male Characters Good Characters Blue Eyes Silver Hair Unclassified 81
Female Characters Good Characters Blue Eyes Strawberry Blond Hair Pansexual Characters 74
Male Characters Good Characters Blue Eyes Red Hair Homosexual Characters 62

Top Predicted Characteristics (By Sex)

Summary Stats for Predicted Appearances (By Sex)
SEX Predicted # of Appearances # of Characters
Genderfluid Characters 29 1
Female Characters 7 766
Agender Characters 6 14
Male Characters 5 2311

Top Predicted Characteristics (By Alignment)

Summary Stats for Predicted Appearances (By Good/Bad Alignment)
ALIGN Predicted # of Appearances # of Characters
Good Characters 8 1058
Neutral Characters 5 499
Bad Characters 4 1535

Top Predicted Characteristics (By Eye)

Summary Stats for Predicted Appearances (By Eye)
EYE Predicted # of Appearances # of Characters
Hazel Eyes 18 16
Variable Eyes 13 12
Blue Eyes 11 434
Amber Eyes 10 2
Grey Eyes 10 24
Brown Eyes 9 382

Top Predicted Characteristics (By Hair)

Summary Stats for Predicted Appearances (By Hair)
HAIR Predicted # of Appearances # of Characters
Silver Hair 42 6
Strawberry Blond Hair 30 13
Reddish Blond Hair 25 3
Purple Hair 12 8
Gold Hair 11 1
Auburn Hair 10 14

Introducing (New Character 1)

Characteristics

  • Female (7)

  • Good (8)

  • Hazel Eyes (27)*

  • Red Hair (9)

Expected Appearances between 7 and 27

*High volatility

Introducing (New Character 2)

Gradient Boosting Model

  • Female (7)

  • Neutral (5)

  • Blue Eyes (12)

  • Strawberry Blonde Hair (29)

*Medium volatility

Expected Appearances between 5 and 29

Post Credits (Bonus Character)

Post Credits (Bonus Character)

What if we have a few more data points?

  • Demographics of readers

  • Deeper character dynamics

    • Personality traits

  • Character Powers

  • Interaction between storylines (Good vs Bad)

Post Credits (Bonus Character)

Marvel Character X

Questions?